What it is
Momentum is the empirical tendency for stocks that have outperformed over the past 6–12 months to keep outperforming over the next several months — and for laggards to keep lagging. It is a relative signal: you rank the whole universe by trailing return and lean toward the top.
The standard academic construction is "12-1" momentum: rank stocks by their trailing twelve-month price return, but skip the most recent month. The skip matters — over very short windows (days to a few weeks) stocks tend to reverse, not continue, so including last month pollutes the signal with short-term noise.
Momentum = trailing 12-month price return (relative to the universe)
That is the whole idea. No earnings model, no discounted cash flow — momentum asks a single question: is this stock already winning?
Why it works
Momentum is not a quirk of one market or one decade. Jegadeesh and Titman documented it in U.S. stocks in 1993; it has since been found in nearly every equity market studied, in bonds, commodities, and currencies, and across more than a century of data. It is robust enough that Fama and French — long-time skeptics of anything beyond value and size — eventually acknowledged it as a distinct factor.
The leading explanations are behavioral:
- Under-reaction to news. When a company reports genuinely good news, the price often moves part of the way immediately and the rest over the following weeks as the market digests it. The drift is the momentum.
- Gradual information diffusion. Information spreads unevenly — funds, analysts, and retail investors learn at different speeds, so buying pressure arrives in waves rather than all at once.
- Herding and confirmation. Rising prices attract attention, attention attracts buyers, and the trend reinforces itself until something breaks it.
There is also a risk-based reading: momentum pays you for bearing the risk of sharp, occasional crashes (below). Whatever the mechanism, the effect has persisted for decades after publication — which is rare for a market anomaly and suggests it is structural, not a data-mining artifact.
What "good" looks like
There is no single threshold — momentum is relative — but rough intuition for a trailing-12-month figure:
- Negative: A laggard. Avoid for a momentum strategy, however cheap it looks.
- 0–20%: Roughly market-like. Not a momentum name.
- 20–60%: Genuine relative strength. The bulk of a momentum book lives here.
- Above 100%: A market leader in a powerful trend. These are the names momentum is built to ride — but they carry the most reversal risk, so position-sizing and diversification matter more, not less.
The signal is strongest when the move is broad and persistent — a steady year-long climb — rather than a single overnight gap on one headline.
Common gotchas
- Momentum crashes. This is the one that hurts. At sharp market turning points — March 2009, the most violent example — the prior losers rip higher and the prior winners get sold, so a naive momentum book can suffer a brutal, fast drawdown. Momentum's worst months are far worse than its best months are good. Any serious momentum strategy needs diversification and risk controls, not just a ranking.
- Junk rallies. Pure price momentum will happily buy a speculative, cash-burning stock that has tripled on a story. Some of those keep running; many round-trip to zero. Pairing momentum with a quality filter screens out the worst of this (see below).
- Turnover and costs. Momentum decays — a winner today may not be a winner in three months — so the strategy trades more than buy-and-hold. Transaction costs and taxes eat into the paper edge if you rebalance carelessly.
- Crowding. Momentum is well known, and when too much capital chases the same recent winners, the trade can become fragile. Concentration in a single hot theme is the warning sign.
Why the Bull Rankings model doesn't score on it
Momentum is real — but the Bull Rankings model deliberately does not use it as a scoring signal, and that was an evidence-based choice, not an oversight.
The model is a fundamentals-first quality-growth score: it leans toward strong, growing businesses trading at a fair price, which often means buying names that are out of favor rather than already running. When we tested price-momentum and trend overlays on that model, they consistently degraded returns out of sample — because a "buy quality at a fair price" strategy and a "buy what's already winning" strategy pull in opposite directions, and bolting a trend filter onto the former removes exactly the cheap, recovering names that drive its edge. (We wrote up that and other rejected signals in the research behind the rankings.)
So momentum earns its place here as something worth understanding — it's a powerful, well-documented effect — not as an input to the score. If you want to act on it, you'd run it as its own diversified, quality-gated sleeve with real risk controls, exactly because of the crash risk described above; it is not what this site's number measures.
Bottom line
Momentum is the simplest durable edge in markets: winners tend to keep winning for a while. It is robust across centuries and asset classes, it is rooted in how slowly humans price new information — and it is dangerous precisely when markets turn, which is why it must be paired with quality and diversification rather than chased on its own. Buy strong businesses whose stocks are already working, size them sensibly, and respect the reversal risk.
Why momentum is not in our score
Most sites explaining momentum stop at the theory. We tested it, wanted it to work, and left it out — so it's worth saying exactly what happened.
A 6–12 month momentum tilt, a 200-day moving-average filter, and a golden-cross bonus were all tried as inputs to the Bull Rankings score. Every one of them degraded returns out of sample — some by double digits of annual return, with deeper drawdowns than leaving them out.
The reason is structural rather than bad luck, and it's specific to a model like this one. The value pillar tilts the screen toward names trading at a fair or cheap price. Those names are frequently below trend precisely because they're out of favor — that's often why they're cheap. A trend filter systematically avoids exactly the names the rest of the model is trying to find. It doesn't merely fail to add anything; it removes part of where the return comes from.
That's the trap in combining factors that each look good alone. Momentum is a real, well-documented effect. It is also, in this particular model, in direct opposition to the pillar doing most of the work.
What we do use instead
Price behavior isn't ignored entirely — it just isn't a scoring input.
Stock pages show where the price sits in its 52-week range and how far it is off its 12-month high, because that's genuinely useful context when you're reading a company. Knowing that a name scoring well on fundamentals is trading 40% below its high tells you what kind of situation you're looking at. It changes how you'd size it, or whether you'd wait.
But it stays a data point on the page rather than a term in the number. The distinction matters: everything in the score has to earn its place by improving measured outcomes, and momentum was tested against that bar and failed it.
What this means if you're a momentum investor
Nothing here says momentum doesn't work. It says it didn't work here, wired into a fundamentals-and-valuation model that was already capturing part of the same signal.
If you run momentum as a standalone strategy, this site's rankings are close to orthogonal to what you're doing — they will regularly surface names that are cheap and falling, which is the opposite of a momentum entry. Used together, the sensible reading is that a high Bull Rankings score tells you the business is sound and reasonably priced; your own trend rules would tell you when. Those are different questions, and we only answer the first.